Critics of the pre-1914 British economy typically complain about market failure and lack of effective government action to correct it. In the post-1945 context of a mixed economy with an enhanced role for government intervention, the problems increasingly seemed to be of government failure.
‘Government failure’ occurs where the choice or implementation of policy leads to outcomes that are inefficient.7 The standard reasons for government failure include inadequate information, inexperience, principal-agent problems (inability to incentivize officials to work effectively), asymmetric lobbying, inability to make credible commitments, and vote-seeking by politicians. Areas of concern include the structure of taxation, the performance of nationalized industries, industrial policy, competition policy and international trade policy.
The context for post-war taxation was a much increased level of public expenditure. Through to
the mid-1960s public expenditure was around 38 per cent of GDP compared with a pre-war figure of 26 per cent (social = 10.5 per cent of GDP) and then rose to 43 per cent (social = 21.5 per cent of GDP) in 1973. The increase was matched by a similar rise in the tax take which came chiefly from taxes on income. Broadly speaking, the rise in public expenditure was associated with the expansion of the welfare state which was funded by increases in direct taxation. In other words, there was much more redistribution of income. All this was not really surprising in the context of the change in the median voter and shift in the centre of the political spectrum since Edwardian times. Although these changes were instigated by Labour, they were accepted by the Conservatives and they were not vastly out of line with other European countries.
A tax system more conducive to growth would have broadened the tax base, reduced high marginal tax rates, and shifted the balance away from direct to indirect taxation. In this regard, Tanzi (1969) was highly critical of income tax on the grounds that it featured very high marginal rates – the top rate was 97.5 per cent in 1949 and 88.75 per cent in 1973 – he described the system as the least growth friendly of all the countries included in his study. Another notable failure was the long delay before Value Added Tax was finally introduced in 1973. Nominal rates of corporate income taxation were also high – at least 40 per cent in each year and as high as 56.25 per cent at the peak. However, especially in the 1960s and 1970s, effective marginal corporate tax rates were much lower in the light of generous depreciation allowances and high inflation – on average as low as 3.7 per cent by 1980 (King and Fullerton, 1984). The issue here was that large distortions arose through the varying treatment given to different types of investment. Much of this diagnosis was shared by the politicians of the day, especially in the Conservative Party, but such changes were felt to be vote losers and, in particular, to be unwelcome to trade union leaders (Daunton, 2002).
A flagship policy of the Attlee government was the nationalization of a substantial component of the United Kingdom economy including public utilities and transport during the late 1940s. This meant that a sizeable fraction of investment would be undertaken by the state rather than the private sector. In a typical post-war year (1971), the nationalized industries accounted for 18.7 per cent of investment, 7.2 per cent of employment and 10.2 per cent of GDP (Corti, 1976). With the exception of the coal industry, the nationalization of which was a concession to a powerful trade union, there was a market-failure rationale (natural monopoly, externalities) for the enterprises taken into state ownership (Millward, 1997) and, in ignorance of the serious principal-agent problems which would proliferate under state ownership, such actions were favoured by many economists at the time (Shleifer, 1998).
By the 1970s, it was clear that this was an experiment that had failed. The productivity and financial performance of nationalized industries was deeply disappointing. Both inefficient use of labour and excessive investment were serious problems (Vickers and Yarrow, 1988). In 1973 TFP in electricity, gas and water (UK = 100) was 219 in the United States, 119 in West Germany and 88 in France. The figures for labour productivity were 370, 134 and 143, respectively. In every case these productivity gaps far exceed those in manufacturing (O’Mahony, 1999). It became apparent that pricing and investment rules were flouted either by management or through political interference (NEDO, 1976). This amounted to ownership without effective control. The Conservatives did not seriously contemplate privatization, a policy which would have been anathema to the trade union movement.
Over time in the context of pursuing policies which were acceptable to both sides of industry, there was an increasing emphasis on the use of ‘industrial policy’ to promote faster growth.
‘Industrial policy’ aims to change the distribution of resources across economic sectors and activities (Warwick, 2013). Thus, it includes both ‘horizontal’ policies which focus on activities such as innovation, investment and education, while ‘selective’ policies aim to increase the size of particular sectors. Selective policies may try to help struggling sectors faced with downsizing or may seek to develop new competitive advantages. The classic justification for industrial policy is that it remedies market failures, for example, by providing public goods, solving coordination problems or subsidizing activities with positive externalities. More generally, the development of endogenous- growth theory suggests that horizontal policies which raise the appropriable rate of return to innovation and/or investment can have positive effects on the rate of growth. On the other hand, as British experience underlines, there is a high risk of government failure as decisions are distorted by rent-and vote-seeking.
Throughout the period, policies were adopted to enhance investment-led growth with a variety of subsidies to investment in the form of depreciation allowances or grants. At their peak in 1978 these subsidies amounted to 10 per cent of fixed investment but they are widely thought to have been a badly designed policy which was poorly targeted. The econometric evidence is that they had little effect on the volume of investment over the long run (Sumner, 1999) with the implication that there was a large deadweight cost.
In addition, selective policies were used increasingly over time but also with little success.
Although ‘picking winners’ may have been the aspiration, “it was losers like Rolls Royce, British Leyland and Alfred Herbert who picked Ministers” (Morris and Stout, 1985, p. 873). There was a
very clear tendency for selective subsidies to be skewed towards relatively few industries, notably aircraft, shipbuilding and, latterly, motor vehicles (Wren, 1996a). More generally, there was quite a strong bias towards shoring up ailing industries which is well reflected in the portfolio of holdings of the National Enterprise Board (Wren, 1996b), in the pattern of tariff protection across sectors (Greenaway and Milner, 1994), and in the nationalization of British Leyland.8 Moreover, policies to subsidize British high-technology industries were notably unsuccessful in this period in a number of cases including civil aircraft, which by 1974 had cost £1.5 billion at 1974 prices for a return of £0.14 billion (Gardner, 1976), computers (Hendry, 1989) and nuclear power (Cowan, 1990).9 A horizontal policy such as an R & D tax credit would surely have been more appropriate than vain attempts to create ‘national champions’.
A key feature of the Golden-Age British economy was the weakness of competition in product markets which had developed in the 1930s and intensified subsequently. Competition policy was largely ineffective and market power was substantial. Competition policy was inaugurated with the Monopolies and Restrictive Practices Commission in 1948, evolved through the Restrictive Practices Act (1956) and the Monopolies and Mergers Commission (1965), but was mostly ineffective (Clarke et al., 1998). Few investigations took place, very few mergers were prevented, the process was politicized, a variety of ‘public-interest’ defences for anti-competitive activities were allowed, and there were no penalties for bad behaviour. Only the Restrictive Practices Act had teeth but its attack on collusion was ultimately undermined by cartels being superseded by mergers. Competition policy was not seen by economists at this time as important for productivity performance (Broadberry and Crafts, 2001) while neither big business nor the trade union movement had any great enthusiasm for aggressive measures which might threaten their supernormal profits and wages (Mercer, 1995).
Not surprisingly, there is evidence that the British economy was characterized by substantial market power in this period. Initially, collusive activity was widespread; an examination of the agreements registered in compliance with the 1956 Act shows that only 27 per cent of manufacturing was free of price-fixing and 35.7 per cent was cartelized (Broadberry and Crafts, 2001). Over time, industrial concentration increased steadily such that the average 3-firm concentration ratio across manufacturing sectors was 41 per cent by 1968 compared with 26 per cent in 1935 (Clarke, 1985).
Crafts and Mills (2005) estimated that the price-cost margin in United Kingdom manufacturing during 1954–1973 averaged over 2 compared with around 1.1 in West Germany which is consistent with the finding in Geroski and Jacquemin (1988) that the magnitude and persistence of supernormal profits for large firms during 1949 to 1977 was large in the United Kingdom but that significant deviations
from competitive outcomes were not observed in West Germany in the 1960s and 1970s.10
The evidence is that weak competition had an adverse effect on United Kingdom productivity performance during the Golden Age. Broadberry and Crafts (1996) found that cartelization was strongly negatively related to productivity growth in a cross section of manufacturing industries for 1954–63. This result is borne out by the difference-in-differences analysis in Symeonidis (2008) who showed that when cartels were abandoned following the 1956 Restrictive Practices Act labour productivity growth in formerly-colluding sectors rose by 1.8 percentage points per year in 1964–1973 compared with 1954–1963. This finding suggests that a more vigorous competition policy would have improved productivity performance.
Weak competition in product markets was buttressed by protectionism which the United Kingdom was very slow to give up by comparison with its European peer group. Delayed entry to the EEC which the United Kingdom did not join until 1973 was an important part of this. Accordingly, average tariff rates on United Kingdom manufactures were still 14.5 per cent in 1960 compared with 14.7 per cent in 1935 and a comparison for 1958 showed tariffs were typically 2 to 3 times the rate for the same industrial sector in West Germany. Whereas trade costs fell dramatically within the EEC in the 1960s for the United Kingdom this was delayed until the 1970s (Crafts, 2012, Table 2.5).
Failure to liberalize trade underpinned market power as reflected by high price-cost margins (Hitiris, 1978).
Ex-post analysis suggests that joining the EEC raised the level of United Kingdom GDP by about 8 to 10 per cent notably through increasing the volume of trade and strengthening competition (Crafts, 2016). Realizing this gain earlier would have made relative economic decline during the Golden Age a good deal less acute. However, this outcome is much better than was predicted by economists ex ante; in the early 1970s, the economic case for EEC entry appeared finely balanced. By then, however, there was considerable business support in view of anticipated profits from better market access (Rollings, 2007).
As this discussion has revealed, there were numerous reasons for government failure during the
‘Golden Age’. Nevertheless, there was also a common factor which informed these policies, namely, a desire to not to upset organized labour and, indeed, even a willingness to accept an implicit trade union veto on policy reforms. This makes sense given the electoral importance that was attached to short-run macroeconomic outcomes, especially the imperative of achieving a very low rate of unemployment without igniting inflation. This encouraged successive governments, notably in the 1950s and the 1970s, to seek wage restraint through accepting serious constraints on economic policy
(Flanagan et al., 1983).11 The policy worked initially in the sense that the ‘post-war settlement’
reduced the Non-accelerating Inflation Rate of Unemployment (NAIRU) quite considerably and permitted the very low levels of unemployment seen in the 1950s (Broadberry, 1994) but in the long run there was a significant cost in terms of inferior productivity performance.